The use of artificial intelligence to assess diabetic eye disease among the Greenlandic population.

IF 1.3 4区 医学 Q4 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH International Journal of Circumpolar Health Pub Date : 2024-12-01 Epub Date: 2024-02-15 DOI:10.1080/22423982.2024.2314802
Trine Jul Larsen, Maria Bråthen Pettersen, Helena Nygaard Jensen, Michael Lynge Pedersen, Henrik Lund-Andersen, Marit Eika Jørgensen, Stine Byberg
{"title":"The use of artificial intelligence to assess diabetic eye disease among the Greenlandic population.","authors":"Trine Jul Larsen, Maria Bråthen Pettersen, Helena Nygaard Jensen, Michael Lynge Pedersen, Henrik Lund-Andersen, Marit Eika Jørgensen, Stine Byberg","doi":"10.1080/22423982.2024.2314802","DOIUrl":null,"url":null,"abstract":"<p><p><b>Background:</b> Retina fundus images conducted in Greenland are telemedically assessed for diabetic retinopathy by ophthalmological nurses in Denmark. Applying an AI grading solution, in a Greenlandic setting, could potentially improve the efficiency and cost-effectiveness of DR screening.<b>Method:</b> We developed an AI model using retina fundus photos, performed on persons registered with diabetes in Greenland and Denmark, using Optos® ultra wide-field scanning laser ophthalmoscope, graded according to ICDR.Using the ResNet50 network we compared the model's ability to distinguish between different images of ICDR severity levels in a confusion matrix.<b>Results:</b> Comparing images with ICDR level 0 to images of ICDR level 4 resulted in an accuracy of 0.9655, AUC of 0.9905, sensitivity and specificity of 96.6%.Comparing ICDR levels 0,1,2 with ICDR levels 3,4, we achieved a performance with an accuracy of 0.8077, an AUC of 0.8728, a sensitivity of 84.6% and a specificity of 78.8%. For the other comparisons, we achieved a modest performance.<b>Conclusion:</b> We developed an AI model using Greenlandic data, to automatically detect DR on Optos retina fundus images. The sensitivity and specificity were too low for our model to be applied directly in a clinical setting, thus optimising the model should be prioritised.</p>","PeriodicalId":13930,"journal":{"name":"International Journal of Circumpolar Health","volume":null,"pages":null},"PeriodicalIF":1.3000,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10877649/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Circumpolar Health","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/22423982.2024.2314802","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/2/15 0:00:00","PubModel":"Epub","JCR":"Q4","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0

Abstract

Background: Retina fundus images conducted in Greenland are telemedically assessed for diabetic retinopathy by ophthalmological nurses in Denmark. Applying an AI grading solution, in a Greenlandic setting, could potentially improve the efficiency and cost-effectiveness of DR screening.Method: We developed an AI model using retina fundus photos, performed on persons registered with diabetes in Greenland and Denmark, using Optos® ultra wide-field scanning laser ophthalmoscope, graded according to ICDR.Using the ResNet50 network we compared the model's ability to distinguish between different images of ICDR severity levels in a confusion matrix.Results: Comparing images with ICDR level 0 to images of ICDR level 4 resulted in an accuracy of 0.9655, AUC of 0.9905, sensitivity and specificity of 96.6%.Comparing ICDR levels 0,1,2 with ICDR levels 3,4, we achieved a performance with an accuracy of 0.8077, an AUC of 0.8728, a sensitivity of 84.6% and a specificity of 78.8%. For the other comparisons, we achieved a modest performance.Conclusion: We developed an AI model using Greenlandic data, to automatically detect DR on Optos retina fundus images. The sensitivity and specificity were too low for our model to be applied directly in a clinical setting, thus optimising the model should be prioritised.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
利用人工智能评估格陵兰岛居民的糖尿病眼病。
背景:丹麦的眼科护士在格陵兰岛通过远程医疗方式对视网膜眼底图像进行糖尿病视网膜病变评估。在格陵兰环境中应用人工智能分级解决方案,有可能提高糖尿病视网膜病变筛查的效率和成本效益:我们使用 Optos® 超宽视场扫描激光眼底镜为格陵兰和丹麦的糖尿病登记患者拍摄了视网膜眼底照片,并根据 ICDR 进行了分级,利用 ResNet50 网络开发了一个人工智能模型:将 ICDR 0 级图像与 ICDR 4 级图像进行比较,准确率为 0.9655,AUC 为 0.9905,灵敏度和特异性均为 96.6%。将 ICDR 0、1、2 级与 ICDR 3、4 级进行比较,我们的准确率为 0.8077,AUC 为 0.8728,灵敏度为 84.6%,特异性为 78.8%。在其他比较中,我们的表现一般:我们利用格陵兰岛的数据开发了一个人工智能模型,用于自动检测 Optos 视网膜眼底图像上的 DR。由于灵敏度和特异性太低,我们的模型无法直接应用于临床,因此应优先优化模型。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
International Journal of Circumpolar Health
International Journal of Circumpolar Health 医学-公共卫生、环境卫生与职业卫生
CiteScore
2.10
自引率
15.40%
发文量
51
审稿时长
6-12 weeks
期刊介绍: The International Journal of Circumpolar Health is published by Taylor & Francis on behalf of the Circumpolar Health Research Network [CircHNet]. The journal follows the tradition initiated by its predecessor, Arctic Medical Research. The journal specializes in circumpolar health. It provides a forum for many disciplines, including the biomedical sciences, social sciences, and humanities as they relate to human health in high latitude environments. The journal has a particular interest in the health of indigenous peoples. It is a vehicle for dissemination and exchange of knowledge among researchers, policy makers, practitioners, and those they serve. International Journal of Circumpolar Health welcomes Original Research Articles, Review Articles, Short Communications, Book Reviews, Dissertation Summaries, History and Biography, Clinical Case Reports, Public Health Practice, Conference and Workshop Reports, and Letters to the Editor.
期刊最新文献
Living with chronic obstructive pulmonary disease or type 2 diabetes in Greenland - a qualitative interpretive description study. Community directed assessment of pain in a northern Saskatchewan Cree community. Orientation affects the integrity of glass ampoules of 1 in 1000 adrenaline on exposure to very low temperatures. Ethnic differences in CT derived abdominal body composition measures: a comparative retrospect pilot study between European and Inuit study population. Use of telehealth for paediatric rehabilitation needs of Indigenous children - a scoping review.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1